Identification of weak buses for proper placement of reactive compensation through sensitivity analysis using a neural network surrogate model

The aim of this work is to present a surrogate model appropriate for carrying out sensitivity analysis studies. The surrogate model is obtained using a radial basis functions neural network. The study is based on the sensitivity of the overall power system bus voltage magnitudes to reactive power change. The objective is to locate a suitable power system bus for reactive compensation. To validate the proposed method, a power flow based sensitivity analysis is carried out in a power system in order to identify the most vulnerable bus. The findings can be used to identify buses where reactive power compensation can have the most impact. Simulation results are presented that confirms the validity of the proposal.